GA1LLATI.N NATIONAL FOREST

1990 SOIL MONITORING PROGRAM

SUMMARY REPORT .A.Y " .

by Henry F. Shovic and Greg Widner :. .

Soil and Watershed Section

Gallatin National Forest

U. S. Forest Service . ' . '., , . , , .

Bozeman, MT

. ., , + - ". . . . ,

, ,

February 28,

Introduction

One of the basic timber management assumptions used in the Forest Service is that of "sustainability" or the maintenance of long term productivion. The concept of sustained yield includes the maintenance or increase of site productivity. Recently, however, scientists have begun to question whether this is a valid assumption. Recent evidence suggests that in some instances, productivity decreases in intensively managed forests. Also, the unprecedented changes in global climate will certainly increase stress on mountain ecosystems, as well as change their productivity. This makes it critical to understand and preserve the non-climate portions of those ecosystems.

The long term sustainability of forest and range ecosystems depends heavily on maintaining soil productivity. Soil, along with climate, physiography, and biology sets the limits on site productivity within a region through its control of nutrients, moisture, and air supplies to plant roots. Since management influences are largely confined to soil manipulation, the soil condition is a good indicator of the status of land productivity.

Much work has shown the critical importance of the surface layer (A horizon) in maintaining healthy ecosystems. The surface soil is slow to develop, slow to recover, but also slow to reflect natural changes. In mountain enviroaments this surface layer is where most biological activity takes place, and cannot be replaced or even approximated by deeper layers.

Soil properties that greatly influence plant growth are: soil volume, soil porosity, water availability, chemistry, biology, and ecological diversity. Changes in these properties changes plant growth potential. These properties are indicated by soil parameters such as bulk density (level of compaction), organic matter, cation exchange capacity (fertility), base saturation, texture, Ph, and evidence of plant roots. Table 1. indicates some of these parameters for a typical soil on the Gallatin National Forest. Note that there are large differences in parameters (and associated properties) between the surface layer (A horizon), and lower layers. Disturbance of the surface layer correlates with changes in soil properties, and a loss of or detrimental change in this layer has been shown to result in a reduction in productivity.

Table 1. Soil parameters for typical soils on the Gallatin National Forest,

Research efforts are under way to establish better relationships between soil disturbance and site productivity. Regional standards (given *below) have been deveIoped for Region Six and One, and establish quantitative criteria relating to the current level of knowledge. Though these are broad and somewhat liberal, they are the best approximation available.

Soil Homoa

A (surface layer)

B (subsoil)

C (substratum)

Definitions

Site productivity is the inherent capacity of the land to produce wood. This is fixed by local climatic, topographic, and edaphic (soil) conditions, and is expressed through the genetic growth capacity of forest trees.

Organic Matter

4.2

12

0.7

Soil productivity is the soil component of site productivity. It is b e d in the short term (over a rotation), but can vary over 1000's of years. Soil quality is the inberent productivity of the soil. It is measured by certain properties that influence plant growth. Soil pore space/volume relations, surface soil depth, aeration, nutrient supplying ability are important properties measured below. Significant management induced changes in these properties will affect long term site productivity.

Objectives

Cation Exchange Capacity (mcq/lOO g)

24

14

18

We do three general kinds of monitoring relating to the forest plan. These are implementation, effectiveness, and validation. This project is in the "effectiveness" category. Given we are implementing our best management practices (BMP's), we want to know how effective they are at protecting soil quality. Future work will be concentrated in the "validation" category, where we want to know how soil disturbance will affect soil and site productivity.

Specifically, this project is to evaluate soil disturbance in clearcut harvest units. System roads are excluded. Selection criteria for sample sales are 1) used for BMP reviews in

Sand (%)

42

54

61

Silt (%)

38

30

23

Clay (%)

20

16

16

h k Prag- men& (%)

IS

M

65

Roots

many fine; many medium; many coarse; ftw very MBfSe

ftw foarsc; fcar &ry corn=

fcw very came

the last two years, 2) harvested by ground skidding methods, 3) representative of the Gallatin sale program, 4) on a variety of soil types, and 5 ) recently harvested.

Methods

Sampling System

Methods were adapted from "Guidelines for sampling some physical conditions of suflace soils; Howes, S, J. Hazard, and J. Geist, Pacific Northwest Region, July 1983, (R6-RWM- 146-1983). A randomly oriented grid with randomly oriented transects originating from each grid intersection was laid out on the sale map. The number of transects and grid spacing were determined by the unit's area and the precision desired; these calculations are discussed in the statistics section. Transect length was 100 feet, as measurements on this length interval convert directly to percentages. Random azimuths for the grid and transects were obtained by generating a series of random numbers between 0 and 1, then multiplying by 360. Random numbers were generated by an HP-28s handheld calculator with a built in random number generator.

In the field, the starting grid point was arrived at by pacing a distance and direction (measured from the grid map) from some corner or edge of the unit using a 100' tape and handheld compass. The remainder of the grid was laid out in the field in a similar manner using the compass and pacing the distance between grid points. Transects were laid out from each grid intersection at the random azimuths using the compass and 100' tape. The resulting grid over the unit often appeared to be somewhat distorted, but more accurate surveying was not warranted. In any case, the randomness with which the grid and transects were laid out on the ground were not affected by the accuracy of the grid.

Field classification of soil surface condition

Soil surface conditions along each 100' transect were placed in one of seven visual categories at a resolution of one foot, These categories are:

1.) Undisturbed (UN)- this soil's mineral surface and subsurface horizons are in their natural state, or, if equipment has been on the site, has not been disturbed.

2.) Deposited (DE)- this soil has accumulated away from its original location either by equipment or erosion.

3.) Compaction Observed (C0)- this soil is observed to be compacted, but is otherwise undisturbed and the organic layer is intact.

4,) Displaced (D1)- this soil has been removed to another location by equipment or logs.

5.) Eroded (ER)- this soil has been detached and transported by wind, water, ice, or gravity.

6.) Burned (BD)- this soil has been charred.

7.) Puddled (PU)- this soil has been smeared and compacted, usually be equipment or log movement. It is also usually displaced or deposited.

For the purposes of this study, each of these classes are exclusive; any given portion of a transect may be assigned to only one class.

For each class except UN, the amount of vertical displacement is measured relative to the estimated original soil surface and noted. For the CO, DI, ER, and PU classes, this measurement is downward and recorded as a negative number. Vertical displacement is recorded as a positive number in the DE class.

Average depth of disturbance was estimated by using a ruler and tape. A few measurements were taken over a small transect length, and averaged. Measurements were taken to the nearest inch, and transect distance to the nearest foot.

Detrimental and Severe Soil Disturbance

Soil surface condition class alone is not used alone as aa index of soil damage. Forest Service Handbook FSH 2509.18, Soil Management Handbook, outlines the criteria for detrimental changes soil properties and was used here with minor modifications and additions. For this study, the following criteria were used to assess damage in each category:

1.) UN- undisturbed soils are not considered damaged,

2.) DE- deposited soils are' not considered damaged at this time.

3.) CO- compacted soils are considered damaged only if the vertical displacement exceeds one inch. This threshold was developed by assuming a 6" depth affected layer, and a 15% allowable decrease in soil volume over this depth. This is preliminary and will be measured more accurately in the future.

4.) DI- displaced soils are considered damaged if the lesser of 3 inches or 50% of the topsoil is removed. With an average topsoil depth of 9 inches, removal of 3 or more inches of topsoil constitutes damage.

5.) ER- eroded soils are considered damaged if the lesser of 3 inches or 50% of the topsoil has been removed.

6.) BD- burned soils are considered damaged if charring occurs at depths greater than 5 cm (2 in).

7.) PU- all puddled soils are considered damaged.

Total detrimental disturbance is calculated as the sum of the damaged soils in each surface condition class and expressed as a proportion of the total. Total severe disturbance is only the sum of detrimentally displaced soils, as topsoil cannot be replaced within the planning period (100 years).

Statistics Sample Sire and Spacing

For statistical reliability, a sampling density of 1 point per 2 acres was used. From 15 17 transects were used in each sample unit. Grid spacing with the formula:

I = (A/n)^(l/2), where: A= area in square feet n= sample size I = spacing interval

Tests of Statistical Significlvlce (T tests)

Tests of statistical significance (T tests) were performed to establish the "reality" of results. The T test is performed as follows: If t (calculated) > or = to t (tabular), where

t (calculated) = (TDD - 30)/(v(Pi)/n)~' t (tabular) = alpha/2, n-1 df

TDD = Total Detrimental ~&turbance 30 = Regional standard

v(Pi) = sample variance n = sample size then the standard has been exceeded (at an alpha level of 0.2).

Results

Sample Harvest Unit Descriptions

. The Bear-Chestnut timber sale is located on the Bozeman District of the Gallatin National Forest. Unit 13 is located approximately at SE1/4 Sec 10 R7E T3S at an elevation of about 6800 feet ASL Aspect is NNE and slopes are generally in the 10%-30% range. Soils occurring in the unit are classed in soil map units 84-1A (NW2/3 of unit) and 86-2A (SW1/3 of unit) as described in the Gallatin National Forest Soil Suwey. This unit appeared to be in "good shape" after clear-cutting.

The Cooke City Salvage timber sale is located on the Gardiner District of the Gallatin National Forest. Unit 4 is locate approximately at SW1/4 Sec 19 T9S Rl5E at an elevation of about 8400 feet ASL. Aspect is S-SE and slopes are in the 10%-25% range. Soils found on the unit are classed in map unit 71-1C. This unit, as part of a salvage sale, was entirely burned during the Yellowstone fires of 1988. As such, this unit is substantially larger than other units in this survey.

The East Boulder timber sale is located on the Big Timber District of the Gallatin National Forest. Unit 8 is located at approximately SW1/4 Sec 10 T4S R13E at an elevation of about 7500 feet ASL. Aspect is N and slopes are in the 15%-25% range. Soils in the sample unit are classed in map unit 34-1C. This unit appeared to be in particularly difficult terrain as far as ground skidding is concerned. Both live and ephemeral drainages were used as skid roads, while much of the rockier terrain was left untouched by fallers and skidders.

The Iron Mountain timber sale is located on the Big Timber District of the Gallatin National Forest Unit 13 is located at approximately SW1/4 Sec 10 T3S R14W at an elevation of about 6600 feet ASL. Slopes are in the 30% range, and aspect is E-SE. Soils occurring in the unit are classed in map unit 54-5C. This unit appeared steep with regard to ground skidding, yet severe disturbance appeared limited to localized areas.

The Portal Salvage timber sale is located on the Bozeman District of the Gallatin National Forest. Unit 3 is located at approximately SE1/4 Sec 20 T6S M E at about 7300 feet ASL Slopes are generally in the 20% range, and aspect is southwest. Soils occurring in the Unit are classed in map unit 71-2B. This unit contained a designated streamcourse (included in the monitoring) and a permanent road (not included the monitoring). The sale operator scarified the entire unit with a D-7 prior to soil monitoring. It seems plausible that these scarification efforts significantly increased soil damage above that caused by felling and skidding. Also, fallers returned to drop additional timber after the scarification was completed and as the soil monitoring field work was nearing completion.

The West Hebgen timber sale is located on the Hebgea Lake District of the Gallatin National Forest. Unit 26 is located at approximately N W 114 Sec 31 T13S R44E at an elevation of about 7600 feet ASL. Slopes are in the 20%-25% range, and aspect is east. Soils in this unit are classed in map units 54-1G. Assessing the degree of soil disturbance was complicated due to weeds and grasses were well established on the site.

Table 2 summarizes properties of selected sales and harvest units. One objective of this study was to represent the range of conditions occuring on the Gallatin National Forest. The sample sites represent four of five ranger districts, a wide range in size, the elevation range of most commercial timber land on the Gallatin, the slope range of most tractor sales, and recent sales. Soils texture ranges born moderately fine to moderately coarse. All sales passed BMP implementation reviews in 1989 or 1990.

Table 2. Harvest Unit Parametem (all clearcut, tractor logged) for moni to~d units.

District

Bear Chestnut Bozcman

Imn Mountain 1 Big 'Iimber

Poml Salvage Boztrnnn

Elevation E I I I I

mod. fine; few C. I.

mcd. - mod. tine; many c. f.

15.25% mod cozmc; many c. f-

lMO% 1 E8-90 1 medium; lo. - many e. f. 11 1 I

1&2!% 86-90 mod. fine; few c. f. 11 -

U S % &%9 mod coarse; few - many c. f.

Transect Data

Data for each sample site are given in Table 3. The raw data in each class was filtered using regional standards for "detrimental disturbance" and severe disturbance" defined above. These data were subjected to the T test ta determine whether Regional standards were exceeded for total detrimental disturbance. No statistical tests were run on severe disturbance because values were all well below the cutoff of 15%.

Table 4. gives a different perspective on the transect data Soil removal or deposition in the first two columns are calculated as a proportion of the area affected by each. Therefore they cannot be compared across sales. This shows that, when compared to those given as proportion of the entire sale area, disturbed areas are characterized as having a relatively small extent with a large disturbance. For example, harvest activities in the Bear Chestnut unit have either removed or compacted an average of three inches of soil over the entire disturbed area. This is about 113 of the average surface soil

depth. When calculated as proportion of the entire sale area, (Table 2) an equivalent average of 0.7 inch of surface soil has been removed or compacted over the entire sale area. Only 0.1 inch has been deposited, leaving a net loss of 0.6 inch.

Table 4. Average Vertical Displacement by Hamst Unit

AS Proportion of Disturbed Area

Hamst Unit Sod Removal or Corn- Sod Dcpositim (in) &on (in)

lron Mountain -29 3.7 I I 1

Portal Salvage -20 I 3.0 1 1 I -

West Hcbmn I -1.8 I I 3.2 1

kr Proportion of Entire URI

Soil Removal or Com- paction (in)

Soil Deposition

Figure 1. graphically depicts the raw data before filtering. Figure 2. portrays that information after the "detrimental" and "severe" filter. Note that the totals are less in the latter. Disturbance data were retained only if there is a probable effect on long term productivity.

Figure 1. Transect Data f o r Harvest Un l ts

I

Cooke $a l vage Iron tkuntal n - west Aebgen

k p a s i ted Compacted 0 i sp l aced (ZZ] P d d l ed Burned or Eroded

F i gure 2 . Severe and Detr i menta l Olsturbance by W v e s t Unit

0 40

35

30

25

20

I 5

10

5

0 I PortaI*hivage I l ron Mounta I n W s t HObgQn

Severely ~ l s t u r b e d m Oeulmental Olot.

Conclusions and Recommendations

There were significant detrimental soil effects on all sample sites. Three of the six exceeded Regional standards. An average of 113 of all sites was detrimentally disturbed. This relatively high disturbance is surprising. No attempt is made to separate effects of harvest born those of site preparation. Methods used were preliminary and variabili~ is high. Therefore limited use should be made of the data born any one sample site. However, the methods appear to be appropriate for measurement of disturbance in forested environments and the following conclusions are warranted.

1. Soil Disturbance is high on recent tractor sales on the Gallatin Forest. Disturbance is higher than expected, since these sales represent relatively current practices.

2. Regional standards have been exceeded in one-half the cases, by a statistically significant amount. There is significant "de trhental" and some "severe" disturbance. There has been a significant decrease in soil quality.

3. Long term soil productivity has very probably been affected, as well as long term site productivity.

Changes in harvest and site preparation practices may be needed However, with the present accuracy of data it is difEicult to tell what kind and amount should change. A more detailed monitoring effort should be completed in N 1991 to better d e h e the problem and eliminate potential errors in methods. This project should concentrate on reducing variability, establish the kinds of practices causing the problem, measure compaction with standard methods, and select a wider range of site/soiI combinations.

Finally, other resource uses should be monitored. Trails, especially in the wilderness are both over-used and under-maintained. Range allotments have areas of compacted, smeared soils, especially near streams. We should make an effort to systematically ,

monitor these areas as those activities also affect long term soil productivity.